PubMed HealthSearch

SEARCH · PubMed Health

Results for “Generative models”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 145 records · Page 8Linked to original sources

Identification models of the nervous system.

It has been widely observed that when artificial neural networks are trained by supervised learning to do computations that also occur in the nervous system, the behavior of the model neurons often closely resembles that of the real neurons involved in the task. It is not immediately clear why this should be the case or what use can be made of models generated by supervised learning. Here, recent developments are reviewed and analysed in an attempt to clarify these issues. This analysis is facilitated by treating supervised learning models of the brain as a special case of system identification, a general and well-studied modeling paradigm. The neural systems identification paradigm provides a systematic way to generate realistic models starting with a high-level description of a hypothesized computation and some architectural and physiological constraints about the area being modeled. There is no inherent limitation to the realism that can be incorporated into identification models. This approach eliminates the need to find neural implementation algorithms by ad hoc means and provides neuroscientists with a convenient way to build models that account for observed data.

Algorithms

Simulation model for thrombin generation in plasma.

A simulation model for the production of thrombin in plasma is presented. Values of the reaction rate constants as determined in purified systems are used and the model is tested by comparison of simulations of factor Xa, factor Va and thrombin generation curves with experimental data obtained in thromboplastin-activated plasma. Simulations of the effect of hirudin indicate that factor V is predominantly activated by thrombin and not by factor Xa. The model predicts a threshold value for the factor Xa production which, if exceeded, results in explosive and complete activation of prothrombinase. The dependence of this threshold value on different negative feedback reactions, e.g. the inactivation of thrombin and factor Xa by antithrombin III (+ heparin), is investigated. The threshold value, for control plasma in the range of 1-10 pM total factor Xa production, can be raised two orders of magnitude by accelerated inactivation of factor Xa and prothrombinase but is hardly affected by a tenfold increase in the rate of thrombin inactivation or by increased production of activated protein C. This latter effect, however, results in a more gradual input-response relation between factor Xa input and the extent of prothrombinase activation.

Computer Simulation

Applications of computerized interactive morphometry in pathology. II. A model for computer generated diagnosis.

We present a model for the analysis and tentative diagnosis of pathologic problems by a trained observer utilizing data generated by a video based computerized interactive morphometry system in conjunction with multivariate methods of discriminant classificatory analysis that separate classes based on unweighted numerical values and with adhoc algorithms based on hierarchical analysis. The model was tested with two diagnostic problems that could benefit from a morphometric approach: the classification of non-Hodgkin's lymphomas from routine histologic slides and the distinction of malignant mesotheliomas from benign effusions in smears prepared from pleural effusions. The touch-sensitive screen of the computerized interactive morphometry system enables trained observers to measure the real time image of profiles of interest either by using graphic standards or simply by touching the two extreme points of the diameter of interest. This procedure generates multiple data in the form of classes that can be effectively used toward the more objective discrimination of lesions of difficult classification by analysis with a combination of adhoc diagnostic algorithms and multivariate statistical methods.

Computers

The effects of consultation style on consultee productivity.

This study examined the effectiveness of three different consultation styles adapted from Bindman's typology. Consultees were nurses on eight wards in a state hospital for the retarded, who were assigned to Expert, Resource, and Process consultation groups plus a no-treatment control. Data on the number of new programs independently initiated by consultees were collected during a 6-week base line, 12-week consultation, and 6-week follow-up period. Results showed a general increase in number of programs initiated during the second half of the consultation period, with trends established there continued through the follow-up. Degree of change was directly related to the style of consultation: the Expert role proved no better than the control condition; the Resource and Process roles generated significant consultee activity, with the Process model generating the most programs in both experimental and follow-up periods.

Behavior Therapy

Creating phyllotaxis: the stack-and-drag model.

The genesis of phyllotaxis, the origin of the pattern of appendages on the surface just below the apical extreme of many plants, is an old unsolved puzzle. Whereas many models generate helices, the stack-and-drag model is the first to achieve this in an integral construction from seed to flower. Combination of the principle of gnomonic growth, where consecutive additions have comparable positions, with a "dragging" principle, where the developing zone follows the apical tip, provides a powerful tool in simulating a wide range of phyllotactic manifestations. The influence of three vital parameters for primordial size, compressibility, and canalization (or annular arrangement) helps in understanding the problem's nature.

Mathematics

Comparative molecular modeling of Amphioxus calcium vector protein with calmodulin and troponin C.

Calcium vector protein (CaVP), a new protein isolated from Amphioxus muscle, binds in a Ca2(+)-regulated manner to a 27 kd target protein, named CaVPT, whose function has not been elucidated yet. CaVP bears significant sequence homology to both calmodulin and skeletal muscle troponin C, especially in the C-terminal half of the molecule, which presumably contains the two functional Ca2(+)-binding sites. The N-terminal half contains two abortive EF-hands and is intramolecularly crosslinked with a disulfide bond. Using the crystallographic structures of calmodulin and striated muscle troponin C as a framework, we constructed two different three-dimensional models of CaVP and modeled the intramolecular disulfide bridge. The modeling based upon the coordinates of calmodulin yields a Ca2(+)-filled sites configuration in the N-terminal half of the molecule, even though no Ca2+ is bound in this half, whereas the troponin C-derived model generates a Ca2(+)-empty sites configuration. The models predict that neither is the Ca2(+)-filled nor in the Ca2(+)-empty sites conformation is there any steric and/or energetic obstacle for the formation of the disulfide bridge and that the disulfide bond is poorly accessible to reducing reagents. The optical properties of the Trp and Tyr residues of CaVP indicate that the calmodulin-derived model represents the most plausible prediction.

Amino Acid Sequence

Family therapy in the treatment of adolescents.

A family with an adolescent must transform itself from a predominantly nurturant unit to one that can tolerate and encourage the adolescent's need to separate. When an adolescent presents with symptoms that disrupt the developmental process, the clinician who is familiar with several models of family therapy is better able to select a therapeutic and cost-effective intervention. For family assessments, the authors recommend a biopsychosocial approach, which has replaced the outdated view that families cause psychiatric problems and which acknowledges the family as a potential source of healing. The authors describe several models of family therapy--Satir's communication model, the structural model of Minuchin, Bowen's cross-generational model, and psycho-education--and examine features of these approaches useful for working with adolescents. Illustrative vignettes and some guiding principles for matching model and problem are offered.

Adolescent

The clinical and renal biopsy predictors of long-term outcome in lupus nephritis: a study of 87 patients and review of the literature.

The prognostic markers in 87 consecutive patients with lupus nephritis who underwent renal biopsy are reported for five clinically relevant long-term outcomes--renal insufficiency, renal failure, death due to renal systemic lupus erythematosus, death due to non-renal SLE and death due to SLE, both renal and non-renal. We have demonstrated that a number of previously neglected or rarely studied predictors were important prognostic markers. These included the duration of renal disease before biopsy, overall severity of SLE, as well as the presence of vasculitis, hypertension or a comorbid ailment. Furthermore, the study confirms the predictive importance of serum creatinine, 24-h urinary excretion of protein, C3, and of the activity and chronicity indices on biopsy. However, overall a simple measure of tubulointerstitial disease was the best predictor obtained from biopsy. Prognostic models based on clinical data alone were developed for each of the five outcomes. The models amplify our clinical understanding of lupus nephritis. Markers of renal severity were most important in predicting renal outcomes such as renal insufficiency and renal failure. Prognostic factors less directly related to renal disease (comorbidity and vasculitis) were important predictors of fatality. A marker of immunologic disease activity (C3) was a valuable predictor for many of the outcomes. Thus markers of disease severity reflecting organ damage due to SLE and other comorbid conditions could be combined with markers of immunologic activity to predict a variety of outcomes of relevance to a clinician. When biopsy data obtained by light or electron microscopy were evaluated for their ability to add new predictive information to the clinical models, only a limited value for biopsy was noted. It is likely that this reflected the close correlational relationships between clinical and biopsy variables, the strong clinical models generated, and the inclusion in the clinical models of the previously neglected clinical variables, duration of renal disease before biopsy and the presence of vasculitis or comorbid disease.

Adolescent

Nonword pronunciation and models of word recognition.

Nonword pronunciation is a form of generalization behavior that has been at the center of debates about models of word recognition, the role of rules in explaining behavior, and the adequacy of the parallel distributed processing approach. An experiment yielded data concerning the pronunciation of a large corpus of nonwords. The data were then used to assess 2 models of naming: a model developed by D. C. Plaut and J. L. McClelland (1993), which is similar to the one described by M. S. Seidenberg and J. L. McClelland (1989) but uses improved orthographic and phonological representations, and the grapheme-phoneme correspondence rules of M. Coltheart, B. Curtis, P. Atkins, and M. Haller's (1993) dual-route model. Both models generate plausible nonword pronunciations and match subjects' responses accurately. The dual-route model does so by using rules that generate correct output for most words but mispronounce a significant number of exceptions. The parallel distributed processing model does so by finding a set of weights that allow it to generate correct output for both "rule-governed" items and exceptions. Some ways in which the two approaches differ and other issues facing them are also discussed.

Humans

Expression of N-acetylglucosaminyltransferase III in hepatic nodules generated by different models of rat liver carcinogenesis.

Our earlier studies revealed that N-acetylglucosaminyltransferase III (GnTase III), which catalyzes the insertion of a bisecting N-acetylglucosamine (bi-Gn) in the complex-type N-linked glycans of cellular glycoproteins, is present in hepatic nodules promoted by the orotic acid model. Neither the bi-Gn residues nor the activity of GnTase III is detectable in normal livers and in the surrounding non-nodular liver of the rat. The present study was designed to find out whether an expression of activity of GnTase III is a phenotypic property characteristic of hepatic nodules or simply unique to nodules produced by the orotic acid model. Fischer male 344 rats were initiated with two different carcinogens namely 1,2-dimethyl-hydrazine or diethylnitrosamine and promoted by other models such as the resistant hepatocyte model and the choline deficient diet model, in addition to the orotic acid model. The hepatic nodules generated by these three different models and hepatocellular carcinomas exhibited significant levels of activity of GnTase III while non-nodular surrounding liver, regenerating liver after 2/3 partial hepatectomy or livers of age and sex matched control rats, had no detectable activity. The detection of the activity of GnTase III in nodules and in cancer is in agreement with the presence of bi-Gn residues reported in gamma-glutamyltranspeptidase of cancer tissues of rats. These results are consistent with the conclusion that the expression of GnTase III is activated during hepatocarcinogenesis and is not related to any particular initiator or promoter.

Animals

Biophysical aspects and modelling of ciliary motility.

The dominance of viscous forces in the generation of propulsive thrust by cilia is emphasised. Fourier analysis indicates that ciliary bends consist of circular arcs joined by linear segments; this arc-line shape appears to be a property associated with the molecular mechanism responsible for bending the cilium and is unchanged by variations in the external viscous loading on the organelle. The flexibility of a computer-generated model of axonemal structure is demonstrated by the incorporation of recent data concerning the surface lattice of the microtubules. Computer simulations using the model show that predictions based on stochastic, rather than co-ordinated, dynein arm activity provide a qualitative match to experimental observations of microtubules gliding over fields of dynein molecules.

Animals

A computer model for the 30S ribosome subunit.

We describe a computer-generated model for the locations of the 21 proteins of the 30S subunit of the E. coli ribosome. The model uses a new method of incorporating experimental measurements based on a mathematical technique called distance geometry. In this paper, we use data from two sources: immunoelectron microscopy and neutron-scattering studies. The data are generally self-consistent and lead to a set of relatively well-defined structures in which individual protein coordinates differ by approximately 20 A from one structure to another. Two important features of this calculation are the use of extended proteins rather than just the centers of mass, and the ability to confine the protein locations within an arbitrary boundary surface so that only solutions with an approximate 30S "shape" are permitted.

Computers

Computer modeling 16 S ribosomal RNA.

A three-dimensional structure for 16 S RNA has been produced with a computer protocol that is not dependent on human intervention. This protocol improves upon traditional modeling techniques by using distance geometry to fold the molecule in an objective and reproducible fashion. The method is based on the secondary structure of RNA and treats the molecule as a set of double-stranded helices that are linked by flexible single-strands of variable length. Data derived from chemical cross-linking studies of 16 S RNA and tertiary phylogenetic relationships provide the constraints used to fold the molecule into a compact three-dimensional form. Possibly subjective evaluation of the input data are transformed into verifiable quantitative parameters. Relationships based on general locations within the 30 S subunit or on protein-RNA interactions have been specifically excluded. The resolution of the model exceeds that of electron micrographs and approaches that obtained in preliminary X-ray crystal structures. The model size of 245 x 190 x 140 A is compatible with that of the 30 S subunit as determined by electron microscopy. The volume of the model is 1.87 x 10(6) A which is similar to that of the small subunit in a preliminary X-ray crystal structure. The radius of gyration of the model structure of 76 A is intermediate to that seen for partially denatured and fully folded 16 S RNA. Computer graphics are used to display the results in a manner that maximizes the opportunities for human visual interpretation of the models. A format for displaying the structures has been developed that will make it possible for researchers who have not devoted themselves to ribosomal modeling to comprehend and make use of the information that the models embody. On this basis the computer-generated models are compared with models developed by other researchers and with structural data not included in the folding parameter data set.

Base Sequence

A three-dimensional virtual model of the head generated from digitalized CT or MR images for anatomical-radiological and neurosurgical evaluations.

A computer-generated virtual model has been developed that reconstructs the head in three dimensions from digitalized images obtained with magnetic resonance and computerized tomography. Through programming and the use of commercial graphic animation software, we have developed the whole process of the rotation of the head on different spatial planes. The procedure permits multidirectional anatomical sections to be made on the structure of the head, providing a true dynamic and user-friendly anatomical-radiological atlas. The system serves as a virtual model for the localization of an ideal surgical approach to any lesion thus avoiding possible neurological lesions.

Anatomy, Artistic

A linear time-varying model of force generation in skeletal muscle.

A model of isometric force production by skeletal muscle was developed in which the response to each stimulus in a train was described by a critically damped, linear second-order system. The parameters describing the system were constrained to be constant within an interstimulus interval, but were allowed to vary between interstimulus intervals. The ability of this model to match experimental data, and the time variation in the parameters (low-frequency gain and natural frequency) required to do so were examined in soleus and plantaris muscles of the cat stimulated by synchronous whole-nerve stimulation. The model produced good fits across firing rates from twitch to tetanus for slow and fast muscle, rested and fatigued muscle, and maximal submaximal stimulation. Both gain and natural frequency generally varied smoothly and predictably under all conditions. Gain increased at intermediate stimulation rates and in potentiated muscle, and decreased with fatigue and submaximal stimulation. Natural frequency was higher in fast muscle, and decreased with stimulation rate and fatigue. This modeling approach may provide a useful alternative to current models of skeletal muscle force, as its implementation is simple and it can describe force under conditions (fatigue, potentiation) where the muscle dynamics change with time.

Animals

A model of handwriting.

The research reported here is concerned with hand trajectory planning for the class of movements involved in handwriting. Previous studies show that the kinematics of human two-joint arm movements in the horizontal plane can be described by a model which is based on dynamic minimization of the square of the third derivative of hand position (jerk), integrated over the entire movement. We extend this approach to both the analysis and the synthesis of the trajectories occurring in the generation of handwritten characters. Several basic strokes are identified and possible stroke concatenation rules are suggested. Given a concise symbolic representation of a stroke shape, a simple algorithm computes the complete kinematic specification of the corresponding trajectory. A handwriting generation model based on a kinematics from shape principle and on dynamic optimization is formulated and tested. Good qualitative and quantitative agreement was found between subject recordings and trajectories generated by the model. The simple symbolic representation of hand motion suggested here may permit the central nervous system to learn, store and modify motor action plans for writing in an efficient manner.

Cybernetics

[Approximation to the dynamics of meningococcal meningitis through dynamic systems and time series].

Meningococcal meningitis is subjected to epidemiological surveillance due to its severity and the occasional presentation of epidemic outbreaks. This work analyses previous disease models, generate new ones and analyses monthly cases using ARIMA time series models. The results show that disease dynamics for closed populations is epidemic and the epidemic size is related to the proportion of carriers and the transmissiveness of the agent. In open populations, disease dynamics depends on the admission rate of susceptible and the relative admission of infected individuals. Our model considers a logistic populational growth and carrier admission proportional to populational size, generating an endemic dynamics. Considering a non-instantaneous system response, a greater realism is obtained establishing that the endemic situation may present a dynamics highly sensitive to initial conditions, depending on the transmissiveness and proportion of susceptible individuals in the population. Time series model showed an adequate predictive capacity in terms no longer than 10 months. The lack of long term predictability was attributed to local changes in the proportion of carriers or on transmissiveness that lead to chaotic dynamics over a seasonal pattern. Predictions for 1995 and 1996 were obtained.

Humans

Electromagnetic interference of pacemakers by mobile phones.

The topic of interference of pacemakers by mobile phones has evoked a surprisingly strong interest, not only in pacemaker patients, but also in the public opinion. The latter is the more surprising, as in the past, the problem of interference has scarcely found the attention that it deserves in the interest of the patient. It was the intention of our investigation to test as many pacemaker models as possible to determine whether incompatibility with mobile phones of different modes may exist, using an in vitro measuring setup. We had access to 231 different models of 20 manufacturers. During the measurements, a pulse generator together with a suitable lead was situated in a 0.9 g/L saline solution, and the antenna of a mobile phone was positioned as close as possible. If the pulse generator was disturbed, the antenna was elevated until interference ceased. The gap in which interference occurred was defined as "maximum interference distance." All three nets existing in Germany, the C-net (450 MHz, analogue), the D-net (900 MHz, digital pulsed), and the E-net (1,800 MHz, digital pulsed) were tested in succession. Out of 231 pulse generator models, 103 pieces corresponding to 44.6% were influenced either by C- or D-net, if both results were totaled. However, this view is misleading as no patient will use C- and D-net phones simultaneously. Separated into C- or D-net interference, the result is 30.7% for C or 34.2% for D, respectively, of all models tested. The susceptible models represent 18.6% or 27% of today's living patients, respectively. All models were resistant to the E-net. With respect to D-net phones, all pacemakers of six manufacturers proved to be unaffected. Eleven other manufacturers possessed affected and unaffected models as well. A C-net phone only prolonged up to five pacemaker periods within 10 seconds during dialing without substantial impairment to the patient. Bipolar pacemakers are as susceptible as unipolar ones. The following advice for patients and physicians can be derived from our investigations: though 27% of all patients may have problems with D-net phones (not C- or E-net), the application should generally not be questioned. On the contrary, patients with susceptible devices should be advised that a distance of 20 cm is sufficient to guarantee integrity of the pacemaker with respect to hand held phones. Portables, on the other hand, should have a distance of about 0.5 m. Pacemaker patients really suffering from mobile phones are very rare unless the phone is just positioned in the pocket over the pulse generator. The contralateral pocket or the belt position guarantees, in 99% of all patients, undisturbed operation of the pacemaker. A risk analysis reveals that the portion of patients really suffering from mobile phones is about 1 out of 100,000. Nevertheless, it would be desirable in the future if implanting physicians would use only pacemakers with immunity against mobile phones as guaranteed by the manufacturers.

Electromagnetic Fields